AI Model to Predict Complications After Transcatheter Closure of Perimembranous Ventricular Septal Defect in Children Using Multimodal Clinical Data Integration

What this Trial Is About

Researchers are developing and validating a multimodal artificial intelligence AI model to predict treatment-related complications in children with perimembranous ventricular septal defect pmVSD who undergo transcatheter device closure. This observational study aims to determine if integrating data such as demographics, lab results, electronic health records, echocardiography reports, chest X-rays, and electrocardiograms can help accurately assess individual risk of complications within 30 days after the procedure. The study collects data retrospectively from routine clinical care records of pediatric patients who had the transcatheter closure for pmVSD. Deep learning techniques will be applied to extract features from text and imaging to train and validate the AI prediction model. No direct treatment or intervention is administered as part of this study. Participants involvement consists of the use of their existing medical records and clinical data. Researchers will analyze this information to measure procedure-related complications up to 30 days following closure, and evaluate the models accuracy through metrics like sensitivity, positive and negative predictive values, and precision-recall curves. The study is expected to be completed by the end of 2027.

A Multimodal AI Prediction Model for Complications After Transcatheter Closure of Perimembranous VSD in Children